Susceptibility to and severity of many autoimmune diseases, including type 1 diabetes (T1D), are known to be closely linked with particular class II MHC alleles, but the mechanism of these associations remains unknown. We have found that class II alleles that form unusually low-stability complexes with class II- associated invariant chain peptides (CLIP) are disproportionately represented among alleles that confer susceptibility to autoimmunity. Our recent work indicates that variations in class II/CLIP affinity affect the stability, longevity, and abundance of class II molecules in model antigen presenting cells (APC) and can modulate antigen presentation. Mechanisms by which variations in class II/CLIP affinity could influence susceptibility to autoimmunity thus include alterations in central selection events or peripheral tolerance or activation events, all of which are controlled by antigen presentation. Here, we propose to study whether varying class II/CLIP affinity can influence autoimmune disease pathogenesis in a whole animal. We will: 1) Establish a short-term mouse model of T1D in which the affinity of CLIP for class II has been modulated;this will be achieved by reconstituting irradiated, pre-disease NOD mice with HSC expressing invariant chains with wild type CLIP (with low affinity for MHC II) or mutated CLIP (with high affinity for MHC II);2) Measure the effects of class II/CLIP affinity on class II stability, longevity, and abundance in primary APC types from these mice, including in the context of inflammatory stimuli;3) Determine whether modulation of class II/CLIP affinity in BM-derived APC modulates disease and key immunologic features in this model. The results of these experiments will shape future studies, in which we will extend this work in a long-term model to investigate diabetes pathogenesis using mice that stably express high-affinity CLIP/Ii. If expression of high affinity CLIP in hematopoietic cells is sufficient for disease protection, it will provide new therapeutic options for individuals at risk for T1D. Although it has been known for a number of years that certain proteins, the HLA proteins, are a critical genetic risk factor for type 1 diabetes and other debilitating autoimmune diseases, the explanation for this genetic link has remained unknown. We will perform experiments in a mouse model of diabetes to test a novel hypothesis for the mechanism of this association. If successful, our experiments will shed new light on the mechanism(s) of disease initiation and pathogenesis in diabetes, and may suggest new treatment approaches for people who are at high risk for diabetes.